<html> <head> <base href="https://bugs.freedesktop.org/" /> </head> <body> <div> <a class="bz_bug_link bz_status_NEW " title="NEW - Add FP64 support to the i965 shader backends" href="https://bugs.freedesktop.org/show_bug.cgi?id=92760#c18">Comment # 18</a> on <a class="bz_bug_link bz_status_NEW " title="NEW - Add FP64 support to the i965 shader backends" href="https://bugs.freedesktop.org/show_bug.cgi?id=92760">bug 92760</a> from <a class="email" href="mailto:itoral@igalia.com" title="Iago Toral <itoral@igalia.com>"> Iago Toral</a> <pre>(In reply to Matt Turner from <a href="show_bug.cgi?id=92760#c16">comment #16</a>) > (In reply to Connor Abbott from <a href="show_bug.cgi?id=92760#c15">comment #15</a>) > > 2. "i965/fs: fix ceil/floor for doubles" -- sorry, you're not going to get > > off that easily :) You can't use floats here, since you're going to run into > > range and precision issues. If the piglit tests don't test for those, then > > they need to be fixed. Since we don't have HW support for > > ceil/floor/trunc/etc., you're going to need to implement them using integer > > operations. I was planning on doing this in nir_lower_double_ops.c. I think > > trunc is the easiest to implement using integer ops -- you don't need to > > worry about the sign, you just have to mask off the appropriate number of > > bits in the mantissa based on the exponent -- but the rub is that 64-bit > > integer arithmetic isn't implemented yet in NIR or the backend, and it > > doesn't even exist prior to Broadwell, so you have to split the double into > > 32-bit chunks and operate on those. Then once you've implemented trunc, you > > can implement all the other missing operations in terms of it. Have fun :) > > I knew at one point how to do these... > > You basically implement trunc using bit operations (by looking at the > exponent figure out how many bits of the mantissa you need to mask off, I > think), and then you can implement floor and ceil by trunc(x ± > nextafter(0.5, 0.0)). > > Round-to-even was a little harder and I think required an explicit test > against x.5. > > I'll see if I can dig it back up. Thanks Matt! I had a quick look for this yesterday but did not find exactly what I need yet. If you happen to dig it up back that would be great, I'll keep looking for it on my side though, so no pressure :)</pre> </div> <hr> You are receiving this mail because: <ul> <li>You are the QA Contact for the bug.</li> </ul> </body> </html>